KernelAgent: Hardware-Guided GPU Kernel Optimization via... Kaiming Cheng, Jack Khuu & Laura Wang

Kaiming Cheng, Jack Khuu, and Laura Wang explain how KernelAgent uses GPU hardware signals and multi-agent orchestration to optimize Triton kernels.

Don't miss out! Join us at our next KubeCon + CloudNativeCon event in Salt Lake City, United States (Nov 9–12, 2026). Connect with our current graduated, incubating, and sandbox projects as the community gathers to further the education and advancement of cloud native computing. Learn more at kubecon.io KernelAgent: Hardware-Guided GPU Kernel Optimization via Multi-Agent Orchestration - Kaiming Cheng, Jack Khuu & Laura Wang, Meta Recently, the PyTorch team released KernelAgent, an open agentic system achieving 100% correctness across all 250 L1/L2/L3 KernelBench tasks. We extend that work by adding a hardware-guided optimization layer to the existing framework. Building on the previous correctness-focused pipeline, KernelAgent integrates GPU hardware-performance signals into a closed-loop multi-agent workflow to guide the optimization for Triton Kernels. We evaluate the kernels generated by KernelAgent on all 100 L1 KernelBench tasks. Overall, it achieved 2.02x speedup over generated kernels from earlier versions. On average, KernelAgent generated 1.56x speedup when compared to default torch.compile, outperforming 65 of 100 KernelBench L1 tasks and achieving 89% of the hardware roofline efficiency on the H100. The optimization codebase is located at KernelAgent repo with documentation to get started. We also share a selection of end-to-end KernelAgent optimization artifacts in the open-source repo.